by Joshua Littenberg-Tobias
Computer science education is disproportionately shaped by assumptions and prejudices about who belongs and is represented, creating large disparities by race and gender (Margolis et al., 2008). We developed a simulation on disparities in representation in computer science classes based around a conversation with a reluctant principal who is resisting changing schedules to have more equitable representation in computer science classes. In previous research, we found that many participants failed to recognize the racial/gender imbalance or agreed with the principal when he proposed solutions that did not address the underlying issue (Littenberg-Tobias et al., 2021). In response, we have added an AI intervention within the simulation that responds with feedback when participants’ responses indicate they do not grasp the underlying disparities that cause the differences in representation by race and gender. In this demonstration, we will present the simulation with the AI intervention. We aim to share our contribution of how AI might be used to advance equity in computer science education and get feedback from the AI community.
